EE Seminar: TOP-GAN: Label-Free Cancer Cell Classification Using Deep Learning with a Small Training Set
Speaker: Moran Rubin
M.Sc. student under the supervision of Prof. Natan T. Shaked
Wednesday, November 28th 2018 at 15:30
Room 011, Kitot Bldg., Faculty of Engineering
TOP-GAN: Label-Free Cancer Cell Classification Using Deep Learning with a Small Training Set
Abstract
We propose a deep learning approach for medical imaging that copes with the problem of a small labeled training set, the main bottleneck of deep learning, and apply it for classification of healthy and cancer cells acquired by quantitative phase imaging. The proposed method is hybridization between transfer learning and generative adversarial networks (GANs). Healthy cells and cancer cells of different metastatic potential have been imaged by low-coherence off-axis holography. After the acquisition, the optical path delay maps of the cells have been extracted and directly used as an input to the deep networks. In order to cope with the small number of classified images, we have used the GAN setup to train a large number of unclassified images from another cell type (sperm cells). After this preliminary training, and after transforming the last layers of the network with new ones, we have designed an automatic classifier that copes with a small training set and classified the correct cell type (healthy/primary cancer/metastatic cancer) with 90-99% accuracy. We believe that our approach makes the combination of holographic microscopy and deep learning networks more accessible to the medical field by enabling a rapid, automatic and accurate classification in stain-free imaging flow cytometry.
